Files
solidcraft/docs/render-materials/RIBBON-BLOOM-BATCH-2026-10-23.md
T
9e6e0b8df1 Render workspace: complete experimental ribbon, resources and effects integration (#5493) (#5494)
* wip(render): reconcile completed material and image pipeline source with 0.20 (#5493)

Source-only integration reference for staged review, not a merge-ready aggregate.
Donor: 0xlillith/solidcraft@aea20b0115.
Upstream: 858e5995d6.

Retain original CPU/material contracts and explicit GPU donor provenance in
docs/render-materials-lab.md (#2067, #2191, #2780, #3306). Preserve released
format migrations and current navigation/startup/shell behavior. New saved
records use pending descriptors. The physical-card descriptor follows the
maintainer's material-engineering-cards naming request, with a legacy-lab
format-29 regression. Replace raw new panel hosts with shared containers.

Repository checks, complete workspace/all-target Clippy, and the independent
physical/resource authoring checks pass for this source. The fresh full local
CI invocation has reached test compilation and is not yet complete. Do not
transfer donor native output or prior test totals to this integration.

Generated review screenshots, video, package ZIP and machine logs remain at
the immutable personal evidence pin. Uncommitted geometric displacement is
backed up separately, not silently included. No kernel/vendor change, direct
main write, release or independent contributor-branch rewrite.

Co-authored-by: Christian Svensson <blue@cmd.nu>

* test(cli): pin the Compact camera-sample admission diagnostic (#5493)

The bounded request still refuses 1024x1024x64 samples before host initialization and writes no output. Update only the stale expected diagnostic to the existing Compact budget wording; production limits and code are unchanged. Exact-tree CLI regression suite: 14 passed, 0 failed. The full repository gate has been invoked separately and is not yet complete.

* Normalize UI container violation paths across platforms

Report repository-relative paths with forward slashes so diagnostics and the existing nested-package regression assertions agree on Windows and Unix. Keep the container detection rules unchanged.

Validation: official ui-containers gate and all seven focused xtask tests passed; rustfmt and diff checks passed. Only xtask/dependencies were compiled; no application build or heavy tests.
(cherry picked from commit 44b58d8e10)

* fix(render): preserve camera contracts and correct acceptance regressions (#5493)

Preserve Design view.apply semantics with a separate strict Render camera command. Retain existing translations when English-only aliases collide; add material-axis entries and camera-specific pitch labels. Compare complete validated engineering cards and retain job handles before cancellation.

Focused and native acceptance is pinned to tree f8a423b21f. Full CI has been invoked separately and is not claimed green. Landing follows the render integration policy in #5493.

* fix(render): resolve showcase licence and upstream resource hygiene (#5493)

Record SychPL's explicit MIT OR Apache-2.0 grant, retaining source/model bytes and the Ubuntu lettering notice. Remove 80 absent attribution rows, personal fork records, dead delivery attributes and three fork/resource-only workflows. Existing xtask assets now performs bounded offline hash checks and rejects orphan render attribution; no parallel draft CI or package installation. Original authoring reproduction remains explicit. Scoped xtask tests, assets, Clippy, formatting and prepared SELENE validation pass; no application or renderer-wide result is inferred.

* fix(render): align saved bounds and progressive sample resolution (#5493)

Address review r4364211669/r4364211677/r4364211681/r4364211691/r4364211695. Persisted light/frame domains respect the existing 1e9 document bound; related LED/rectangle/direction cases use the same bound. Workstation primary resolve shares the 4096 admission cap and film counters use checked conversions. Exact candidate: 485 CPU render tests, 21 render document tests and strict scoped Clippy pass, including positive boundary round trips and 2048/4096 end-to-end checkpoints. No released format/migration edits, relaxed test limits or full-stack acceptance claim.

* wip(render): isolate optional resources and harden sample handling (#5493)

Review work in progress: opt-in encoded render library with reproducible runtime catalog exports; native-only robust dependency through the workspace; checked bloom slices; per-sample point-light diagnostics; independent area-light spatial/film keys. Resource-module binaries in both configurations, generation, Cargo production feature graphs and attribution checks pass. Full engine/render regression and combined acceptance are pending: do not merge this donor as a constituent or treat scoped resource proof as runtime acceptance.

* fix(gpu): preserve donor allocations through the scope-safety repair (#5722)

Port the native-validated #5723 scope-cleanup correction into the experimental donor, retaining its adaptive, volume, aging, photographic and bloom allocations. Same reverse-drain helper and regression cases; the donor combination still needs its own compile/test gate and is not covered retroactively by render-base results.

* claim: #1369

* Reject unknown pending concepts after format folding (#1369)

* test(render): close bounded sampling and bloom review regressions (#5493)

Validate the optional-resource/lighting/dependency donor at its exact tree. Required Vulkan GPU 127/0/3; CPU renderer 487/0; engine Render 72/0; document render 21/0; lean library 7/0; strict scoped Clippy, assets/layers/lists/containers and lean-engine WASM pass. Preserve original failed controller/typing attempts. This is the draft donor/disposition source, not a replacement for constituent render PRs or final main acceptance.

* wip(render): preserve stable aging target validation candidate

Exact staged source from prior review work on donor 3b892eab5e. This is a recovery/review checkpoint, NOT accepted delivery. An actual command-path regression currently fails because the selected executor refuses ordered surface aging. No assertion, capability or admission limit is weakened. Native/UI/full combination acceptance remains pending. Saved/runtime split adds one pending descriptor and retains original file scope; no main/render branch is changed.

* test(render): preserve numeric setup intent at the admitted coordinate bound

Follow up jyviko review on #5494 without weakening the corrected saved-coordinate contract. The positive clicked fixture uses 1e9 instead of inadmissible 2e9; explicit out-of-range cases still refuse, and the full saved setup round-trips through JSON and the native container. Exact-source Render workspace 95/0 and strict all-target UI Clippy pass. This does not incorporate the separately failing stable-aging candidate or claim full current-render acceptance.

* test(render): use the real GPU executor for saved aging

Move successful native cold-reopen Check/Start coverage from the CPU-backed protocol fixture into the production GPU executor tests. Verify capture-index changes preserve the stable target and that enabling aging changes both PNG and raw EXR output. Retain explicit, allocation-free refusal coverage for executors without aging support. Production capabilities and admission checks are unchanged.

* fix(render): persist stable identities for targeted aging (#5493)

Separate saved RenderBodyTarget identities from captured mesh ordinals. Reject legacy ambiguous targets, resolve Check/Start after visibility changes and preserve native reopening, copies and undo. Unsupported executors still refuse; real AMD Vulkan regression completes saved aging after reordered capture.

Source tree 6b84c8ad passes document 23, engine Render 75, GPU effects 5 and Render UI 95 with no failures; scoped strict Clippy, WASM and repository gates pass. Fresh CLI/native reopen and clean/aged GPU exports pass with identical physical totals and inspected actual images. Whole current-render integration is separate.

* fix(render): reconcile combined CPU/GPU contracts on current upstream

Unify captured resources/status and CPU-vs-GPU request schemas after the real two-parent integration; remove obsolete duplicate Vulkan selector and retain combined platform-filter regression. Deduplicate only identical or superseded test definitions, retaining the stronger card metadata and reserved-executor assertions. No assertions, admission limits or feature refusals are weakened.

All targets of render dependencies, GPU, engine, UI, CLI and native app pass cargo check with raytrace-cpu and bundled-render-library at tree c70d00d1. Native executable tests and full current-source acceptance remain pending. This is a private integration rehearsal for subsequent constituent PRs, not an aggregate official render/main merge.

* test(render): retain one complete host lifetime regression set

Deduplicate byte-identical DropProbe/task-lifetime tests and retain the adapted environment/texture ownership regressions with every assertion and complete memory estimate. Keep the registered-geometry admission regression once; only struct field order differed. Replace a one-element diagnostic loop with its equivalent block, preserving all wrap/width assertions.

Test-only integration repair. Formatting and diff checks pass; refreshed executable validation is required. No production behavior, limits, assertions or ignores changed.

* fix(doc): recognize the proven format-29 card descriptor alias

Normalize only the historical format-29 physical-material-cards spelling, and only when material-engineering-cards is registered. Admission and migration de-duplication use the same identity. Unknown concepts, wrong versions and incapable readers still refuse before mutation. Six controlled-history Rust regressions and actual CLI migration/no-clobber acceptance pass. No version, released migration or CI changes. Refs #5793, #1369, #5493.

* test(render): preserve grounding refusal before explicit instance moves

Update the two Render binding fixtures for upstream first-occurrence grounding. Verify grounded movement refuses without changing document/history, then explicitly unground through the public command before the intended move. Retain every target, resource, copy and persistence assertion. Refs #5493.

---------

Co-authored-by: Christian Svensson <blue@cmd.nu>
Co-authored-by: Victor Santos Oliveira <56417388+Vittuu@users.noreply.github.com>
Co-authored-by: cfaulkingham <colin.faulkingham@gmail.com>
2026-10-24 14:21:59 +00:00

12 KiB

Render ribbon, native acceptance and display bloom - 2026-10-23

Scope and authority

Owner-directed continuation on the personal lab/render-scale-resources branch, draft PR #7. The source base is 071f5f873bc9d7e98342faf029ad2029c1b98e54. Upstream/main/releases remain untouched. The owner requested a coordinated implementation batch followed by grouped compilation, not a full rebuild after each small edit.

The ribbon follows Nathan Dane/Nateaux's implemented shared shell and the task-order plan in docs/ribbon-layout.md, docs/ui-kit.md, docs/design/ux-redesign/README.md, docs/render-materials/UI_INTEGRATION.md and upstream tracker #2016. Source inspection does not update the frozen integration base or claim completion of that author's separate remaining File-view/layout-preference work.

Implemented source changes, awaiting grouped execution

  • One Render tab with task-ordered Scene, Appearance, Lighting, Environment, Camera and Image groups. Two promoted commands per group, existing SVG artwork, shared section menus, contextual icon accents, command search, pin/customization data and responsive fitting. The Image group receives a bounded reserved area; extra personal pins still use the existing trimming/menu policy rather than overflowing the screen or being deleted.
  • Interactive adapters open the implemented inspector/library for parameterized operations, preserving exact unsaved numeric intent. Nothing is silently applied merely by selecting a library or opening a tool. Direct engine JSON commands retain their existing semantics.
  • Check, Render, Cancel and Export share one action implementation between ribbon, menu/search and sticky footer. The current edited request, active job and exact selected completed result remain authoritative. Disabled reasons reflect the current host/request/result; the engine API is not rewritten around GUI state. Entering Render from Design does not silently launch an expensive default render.
  • A registered render.camera.fit operation frames the current visible subject using the shared camera contract. It preserves orientation and engineering geometry and changes only view state. The generic Design view.apply command is not repurposed.
  • Corrected stale output help that still said EXR was unimplemented. Seven existing Render catalogs receive matching labels and refusal reasons; other supported locales retain the established English fallback.
  • Optional display bloom with explicit strength, relative linear highlight threshold and output-pixel radius. Three normalized separable box passes spread positive highlights with full-kernel normalization and zero outside the image. This is an authored display effect, not measured lens scattering, volumetrics or a transmission-convergence improvement.
  • Bloom advances through bounded cancellable row/column groups, uses fallible allocation, and admits two RGB FLOAT32 scratch images (24 bytes/pixel) before rendering. The GPU raw accumulation and EXR are never modified. Opaque output is required; transparent PNG is explicitly refused rather than flattened. Registered executors must advertise bloom support separately from their older finishing contract.

Presentation order and limits

Raw scene-linear output is retained before denoising and every presentation operation. When enabled, bloom extracts positive highlights from the post-denoise, coverage-premultiplied linear image, convolves them without changing that image, then adds a tone-compressed glow in linear display light after ordinary finishing/background composition and before sRGB encoding. Exposure and creative RGB gains affect the glow. Contrast/saturation/vignette apply to the ordinary presentation; the bloom overlay is the subsequent explicitly creative layer. A zero-strength request has zero scratch cost and unchanged PNG bytes.

Bloom strength is 0..2, threshold 0..1,000,000 relative radiance, and radius 1..64 output pixels. The filter does not claim resolution-independent physical units, energy conservation after artistic threshold/overlay/tone mapping, noise reduction or new transport. The convolution itself has independent direct-kernel, energy, hue, symmetry, chunk-size, invalid-input and cancellation tests.

Native finishing baseline acceptance

The preceding fresh 071f5f87 application is being exercised separately in an isolated profile, with no changes to personal preferences. The actual GUI command pipeline has produced PNG and FLOAT32 EXR outputs at 640x400. A changed presentation and its neutral control produce different PNG bytes and identical raw EXR SHA-256 46cea7fd6091ab16de524a99ae614bb9b7d4b63d1e1c3e8bfb4484eaadc89207. This is real native execution, not a conversion of PNG or a substitute for new ribbon/bloom acceptance.

Provisional captures, receipts and external validation tools are under D:/codex-tools/render-ribbon-20261023. The new batch still requires grouped strict lint/tests, a fresh application and actual light/dark, docked/floating and narrow-window inspection. No green aggregate CI or release is claimed by this recovery checkpoint.

Still separate

Adaptive sample allocation, further transmission-estimator convergence, post-hoc regrading of an old master, volumetrics, arbitrary AOVs/compressed EXR, and remaining advanced material/scene capabilities stay separate roadmap units. Existing unsupported resources retain their availability labels. New ribbon groups and display bloom do not upgrade those concepts to executable capabilities.

Repeated execution exposed a native atomic-publication edge

On the frozen first checkpoint the renderer (465), GPU (100, plus three existing ignores), and Render UI (85) suites passed. The test orchestrator initially selected a nonexistent toolbar::tests:: group; it refused to count zero executed tests as success. After correcting the filter without recompilation, a repeated UI suite observed native Windows access-denied (error 5) while atomically replacing an already created PNG. Twelve isolated repetitions passed, so the external holder/cause of that intermittent denial is not identified.

The existing single-attempt rename is now hardened with an original bounded Windows-only retry around the already-synced rename operation. Only native errors 5/32 are eligible. There are at most eight attempts with requested sleeps totalling 127 ms; all other errors and other platforms keep a single attempt. No target is truncated/deleted, no permission/read-only flag is relaxed, no handle is force-closed, and no security software is disabled. Persistent errors remain errors; the last native code is preserved. The request path scope and explicit overwrite consent are unchanged. Backup publication uses the same narrow rename rule.

Deterministic tests cover exact attempt/wait bounds, permanent/unlisted errors and success. Windows-native tests deliberately open a reader without delete sharing: the original single rename is verified to fail, the old complete bytes stay intact, releasing that reader permits one atomic replacement, and a permanently retained reader still produces an error and leaves the old file untouched. This addresses a reproducible sharing case without claiming that every access-denied event has the same cause.

Only public platform semantics informed this change (no external implementation code):

The Camera group also exposes its implemented numerical inspector directly through the registered, read-only render.camera.get command, alongside Frame and the saved-camera menu action. Direct engine inspection returns an explicit current camera or null rather than inventing a saved view.

Wider guards found inherited locale gaps and a real camera-field loss

The next frozen-source run passed 465 renderer, 85 Render UI, 27 shared ribbon, two pin-state, 16 search, nine atomic-write and five file-scope tests; the remaining document/GPU and shared-kit checks also passed. The intentionally broader localization guard exposed three existing completeness issues: 166 missing Japanese and 253 missing Simplified Chinese Render messages, a missing English registry label for physical backgrounds, and a duplicate English Center X label selecting an untranslated UK key. The complete-catalog contract is retained: the missing Japanese/Chinese messages receive explicit original translations rather than English-filled placeholders or a downgraded completeness test. The physical-background registry label is catalogued, UK center labels are explicit, and the contradictory EXR diagnostic is corrected.

The new orientation test also caught a real earlier bug: explicit view.apply parsed a live Camera through the legacy NamedView shape and silently discarded roll and up. The explicit-camera path now validates and preserves those two optional fields, with unchanged zero-roll/Z-up defaults when omitted. Named-view document storage remains unchanged. Engine tests cover both world-up conventions, malformed-field atomic refusal and legacy defaults; the Render ribbon has an exact orientation-preservation regression. This is a production correction, not relaxation of the failed test.

All source changes in this correction are grouped before the next compilation. Earlier failing receipts remain available and are not relabelled green.

Final grouped code gates and independent feature-profile CI correction

The frozen 106f5d57046ca8854a0dad2671730a9e68aa94d4 / tree eb9ca44845902300b7c2f038c74d4de6bc481216 grouped run completed PASS with source unchanged: 465 renderer tests; 86 Render UI; 27 shared ribbon; two pin-state; 16 command-search; 29 localization; one interactive menu-pin; 69 Render-engine; 17 browser/camera; nine atomic-file; five file-scope; 16 shared-kit guard; 71 shared-widget; 301 document (one explicit existing ignore); and 100 hardware GPU (three explicit existing ignores). A broader workspace substring additionally passed 78 cases; it overlaps other scopes and is not added to a synthetic total. The native application/CLI build and official 12-crate WASM check also passed. Strict workspace/all-target Clippy passed before the run.

Application SHA-256: e09ea35eb164355782c34335129f4be057ecd1a669b22a047246b0d341814211. CLI SHA-256: b1d81d4652d0f0d1dae1b3efcf03c67cefb0722a56f63e1c781b24e36546333e. Real-window acceptance uses this exact immutable build; later test-only/CI/documentation changes are not retroactively attributed to the executable.

Separately, GitHub resource gate 39871253915 failed in both Windows and Linux because the earlier large-primary-film test imported film::primary without its raytrace-cpu feature guard. Workspace feature unification hid this default-profile compilation error locally. The CPU-only test now has the same feature condition as the production module; it is not deleted or disabled when CPU ray tracing is enabled. The resource gate explicitly runs strict lint and unit tests for both default and optional-CPU configurations, with a Python regression requiring that distinction. This is a test/configuration correction with no production-renderer behavior change; the actual default-profile result must be verified separately.

The separately executed default-feature lint then exposed an inherited test-helper ordering lint (CpuExecutor followed the last enabled inline test module only in the no-CPU configuration). That test-only helper is moved ahead of test modules without suppressing the lint or changing runtime code. Both feature profiles are rechecked; a first failure is not a passing matrix result.